1: \documentclass[12pt,epsf]{article}

2: \def\baselinestretch{1.3}

3: \def\dis{\displaystyle}

4: \def\bxi{\mbox{\boldmath$\xi$}}

5: \def\balpha{\mbox{\boldmath$\alpha$}}

6: \def\bbeta{\mbox{\boldmath$\beta$}}

7: \def\bpartial{\mbox{\boldmath$\partial$}}

8: \def\bnabla{\mbox{\boldmath$\nabla$}}

9: \def\bepsilon{\mbox{\boldmath$\epsilon$}}

10: \def\bsigma{\mbox{\boldmath$\sigma$}}

11: \def\bgamma{\mbox{\boldmath$\gamma$}}

12: \def\bpi{\mbox{\boldmath$\pi$}}

13: \def\slM{{\sl M}}

14: \def\bh{{\bf h}}

15: \def\bk{{\bf k}}

16: \def\bp{{\bf p}}

17: \def\br{{\bf r}}

18: \def\bv{{\bf v}}

19: \def\bz{{\bf z}}

20: \def\bx{{\bf x}}

21: \def\by{{\bf y}}

22: \def\bs{{\bf s}}

23: \def\bi{{\bf i}}

24: \def\bj{{\bf j}}

25: \def\bk{{\bf k}}

26: \def\bu{{\bf u}}

27: \def\bL{{\bf L}}

28: \def\bZ{{\bf Z}}

29: \def\bS{{\bf S}}

30: \def\bP{{\bf P}}

31: \def\bX{{\bf X}}

32: \def\bD{{\bf D}}

33: \def\bB{{\bf B}}

34: \def\bJ{{\bf J}}

35: \def\bA{{\bf A}}

36: \def\bE{{\bf E}}

37: \def\bO{{\bf O}}

38: \def\bq{{\bf q}}

39: \def\ba{{\bf a}}

40: \def\bb{{\bf b}}

41: \def\bc{{\bf c}}

42: \def\bC{{\bf C}}

43: \newcommand{\be}{\begin{equation}}

44: \newcommand{\ee}{\end{equation}}

45: \newcommand{\bea}{\begin{eqnarray}}

46: \newcommand{\eea}{\end{eqnarray}}

47: \newcommand{\ra}{\rightarrow}

48: \newcommand{\lra}{\longrightarrow}

49: \newcommand{\lan}{\langle}

50: \newcommand{\ran}{\rangle}

51: \newcommand{\YY}{Y\hspace{-2.75mm}Y}

52: \newcommand{\OO}{\Omega\hspace{-2.35mm}\Omega}

53: \newcommand{\LL}{I\hspace{-1.65mm}L}

54: \newcommand{\PP}{I\hspace{-1.65mm}P}

55: \newcommand{\II}{I\hspace{-1.65mm}I}

56: \newcommand{\hh}{h\hspace{-1.85mm}h}

57: \newcommand{\vp}{v_{\perp}}

58: \newcommand{\gp}{\gamma_{\perp}}

59: \newcommand{\pp}{p_{\perp}}

60: \newcommand{\p}{\partial}

61: \textwidth 15cm

62: \oddsidemargin 8mm

63: \topmargin 0mm

64: \usepackage[dvips]{graphics}

65: 

66: \begin{document}

67: 

68: \vskip 2cm

69: \begin{center}

70: \Large

71: {\bf  Contextual Value-definiteness } \\

72: {\bf and the }\\

73: {\bf Kochen-Specker Paradox } \\

74: \vskip 0.5cm

75: \large

76: Ken Williams \\

77: {\small \sl Department of Physics } \\

78: {\small \sl University of Wisconsin - Parkside } \\

79: \end{center}

80: \thispagestyle{empty}

81: \vskip 0.7cm

82: 

83: \begin{abstract}

84: 

85: Compatibility between the realist tenants of value-definiteness and

86: causality is called into question by several realism impossibility proofs in

87: which their formal elements are shown to conflict. We review how this comes

88: about in the Kochen-Specker and von Neumann proofs and point out a

89: connection between their key assumptions: a constraint on realist causality

90: via additivity in the latter proof, noncontextuality in the former. We

91: conclude that value-definiteness and contextuality are indeed not mutually

92: exclusive.

93: 

94: \end{abstract}

95: 

96: \newpage

97: 

98: \section{overview}

99: 

100: In contrast to Bell's theorem which draws a contradiction between the

101: predictions of quantum mechanics and realism, the theorem of Kochen and

102: Specker (KS), "the second important no-go theory against hidden variable

103: theories", rather calls into question the very logic in realist thinking.

104: The argument is directed against a brand of realism characterized by

105: value-definiteness and noncontextuality, formulated here in section 6.1 as

106: propositions p(1) and p(2), respectively. When these are applied to an

107: elementary QM description of spin-1 particle measurements, a contradiction

108: known as the KS paradox follows.

109: 

110: \noindent In the concluding view of this article, proposition p(2) does not

111: follow from any realist principle and is even anti-realist in the undue

112: constraint it places on causality. Moreover, no realist to the writer's

113: knowledge lays claim to it. As this question concerns the views of realists,

114: it might well be put to rest with a few excerpts from the writings of

115: well-known realist thinkers. But that would fall short of our broader aim to

116: explain how the misunderstanding of a realist p(2) could have come about,

117: and thus point the way to a reasoned resolution. To this end, we consider

118: the two principal interpretations of quantum mechanics, orthodox and

119: realist, with special attention to their contrasting physical

120: interpretations of the QM wave-function; our approach to the KS paradox is

121: within this context.

122: 

123: \noindent We also take the view that the paradox may be well understood from

124: basic quantum mechanics, and so first set about reminding the reader of the

125: relevant elements that he or she may already know from an elementary study

126: of the subject. In section 2 we motivate the discussion of discrete

127: possessed values by considering the quantum signature Stern-Gerlach

128: experimental data, which, in section 3, is described within the framework of

129: quantum mechanics. There, the operational equations and constraints

130: necessary to the KS paradox first appear, equations (\ref{new}). We then

131: consider in section 4 features of the orthodox interpretation relevant to

132: the evaluation of the equations: In 4.1 the orthodox identification of the

133: QM wave-function with individual particle states from which follows in 4.2

134: its view with regard to individual particle possessed-values. Section 5

135: deals with the realist interpretation, there too, particularly regarding

136: features relevant to evaluation of equations (\ref{new}). At the heart of

137: the interpretation is the

138: realist identification of the QM wave function with collections of

139: particles, particle ensembles.

140: 

141: \noindent The KS paradox is derived in section 6 from propositions p(1) and

142: p(2) applied to the set  (\ref{new}). The  connection to the older von

143: Neumann impossibility proof is established in 6.2 and to the later work of

144: Gleason in 6.3. We reconsider the KS paradox in section 7 by close

145: examination of premise p(2). In the end we find little justification for

146: proposition p(2) as a realist tenet, and indeed from the words of realist

147: thinkers, evidence to the contrary. We conclude with an attempt to explain

148: the natural simplicity of contextuality in realist thinking.

149: 

150: 

151: \section{Stern-Gerlach Data}

152: 

153: From the data of Stern-Gerlach (SG) measurements comes a persuasive case

154: that only certain discrete values of intrinsic angular momentum, spin, are

155: observable. There, a stream of unprepared electrons enter and exit a small

156: region of intense inhomogeneous magnetic field strength where it divides in

157: two, half the electrons deflecting up, the other half down by the same set

158: magnitude \cite{101} proportional to a given electron's spin projection

159: along the SG apparatus symmetry axis. Thus measured, each spin projection is

160: found to have one of the two values, $ \pm \frac{1}{2}$ , for which reason

161: the electron is called a spin- $ \frac{1}{2}$ particle. A second measurement

162: along the same SG axis taken on either of the two sub-ensembles invariably

163: confirm the previous result: electrons previously deflected up (down) due to

164: positive (negative) spin of magnitude $ \frac{1}{2}$ along the axis, are

165: again deflected up (down)

166: %

167: \begin{center}

168: \scalebox{1.5}[1.5]{\includegraphics{fig1m.ps}}\\

169: figure 1

170: \end{center}

171: %

172: where $\hat{\bB } $ in figure 1 gives the averaged SG B-field direction. As

173: the pre-measured spin orientations are assumed random, isotropically

174: distributed and uncorrelated to the orientation of the measuring apparatus,

175: this phenomena is unexpected. One would from classical considerations

176: expect the deflection magnitudes to vary continuously with the

177: relative spin-to-apparatus angle

178: %

179: \bea

180: d  \propto s_z & = & s \cos \phi

181: \eea

182: %

183: having maximum and minimum values for $ \phi = 0  \&  \frac{ \pi }{2} $,

184: respectively

185: %

186: \begin{center}

187: \scalebox{1.5}[1.5]{\includegraphics{fig3m.ps}}\\

188: figure 2

189: \end{center}

190: %

191: One finds however that when the apparatus itself is re-oriented (to test

192: against apparatus-to-ensemble correlation), the same result obtains

193: %

194: \begin{center}

195: \scalebox{1.5}[1.5]{\includegraphics{fig6m.ps}}\\

196: figure 3

197: \end{center}

198: %

199: half electrons deflecting up again by the same magnitude as before, half

200: down, only

201: now along the re-oriented axis. Moreover, when this second measurement is

202: taken

203: along an axis perpendicular to the first, again, half are deflected up, half

204: down, by the same magnitude. Indeed, one finds that no matter the SG

205: orientation or the order in which one appears in a sequence of such

206: measurements, the results are always $ + \frac{1}{2} $ or $ - \frac{1}{2}$

207: only . The electron's spin projection is therefore said to be quantized

208: (i.e.,  given in discrete, set amounts) along the measurement axis, called

209: then the axis of quantization, in the figure below the SG

210: symmetry axis specified by $ \theta $ and $\theta^\prime$, SG orientations

211: relative to the laboratory z-axis.

212: %

213: \begin{center}

214: \scalebox{1.5}[1.5]{\includegraphics{fig2mc.ps}}\\

215: figure 4

216: \end{center}

217: %

218: The phenomena poses a difficulty to ordinary intuition. Imagine e.g. that

219: whenever you looked up at the moon its polar tilt (as measured from the

220: plane of sight) varied in

221: $ 180^o $ increments only

222: %

223: \begin{center}

224: \scalebox{0.7}[0.7]{\includegraphics{fig103.ps}}\\

225: \end{center}

226: %

227: no matter the orientation of your head (as measured from the local earth

228: plane).

229: %

230: \begin{center}

231: \scalebox{0.7}[0.7]{\includegraphics{fig104.ps}}\\

232: figure 5

233: \end{center}

234: %

235: The moon's angular momentum would then be said to be quantized.... For the

236: electron this and other unusual

237: microscopic phenomena are well described by the Physics of small-scales,

238: quantum mechanics.

239: 

240: 

241: \section{quantum mechanical description}

242: 

243: Relative frequencies for each of the spin-up, spin-down classes of electron

244: data described above may be obtained from the formalism of quantum

245: mechanics. Experimentally, the frequencies are statistical probabilities

246: over ensembles, collections of identical elements, strictly valid in the

247: large ensemble limit. As regards individual measurements, however, the

248: theory makes no prediction \footnote{ See e.g. ref. \cite[p. \ 31]{103} and

249: ref. \cite[p. \ 2]{102} }. The up (down) deflection probability at apparatus

250: $ \theta $ for a sub-ensemble of electrons previously deflected up at

251: apparatus $ \theta^\prime $ is given by $ \cos^2 \frac{ \phi }{2} $ ( $

252: \sin^2 \frac{

253: \phi }{2} $ ), $ \phi \equiv \theta - \theta^\prime $, assuming for

254: simplicity SG rotation with respect to the laboratory y-axis.

255: %

256: \begin{center}

257: \scalebox{1.5}[1.5]{\includegraphics{fig2mc.ps}}\\

258: figure 6

259: \end{center}

260: %

261: These probabilities are expectation values of projector

262: operators $ {\rm P }_{ \theta \pm} $  in a complex Hilbert space of state

263: vectors $ | \Psi \ran $ whose calculation proceeds as follows:

264: %

265: \begin{equation}

266: \lan  {\rm P }_{ \theta \pm} \ran =  _{\theta^\prime}\lan + | {\rm P }_{

267: \theta \pm}| + \ran_{\theta^\prime} = \left\{ \begin{array}{l} \cos^2 \frac{

268: \phi }{2}     \\  \sin^2 \frac{ \phi }{2} \end{array} \right.  \label{1}

269: \end{equation}

270: %

271: where

272: %

273: \bea

274: |  + \ran_\theta  &=& \cos\frac{\theta}{2} \, | + \ran +  \sin\frac{ \theta

275: }{2} \, | - \ran  \\

276: {\rm P }_{ \theta \pm}  &=&  | \pm \ran_{ \theta }  \, _{\theta }\lan \pm |

277: \label{nnew}

278: \eea

279: %

280: with relations

281: %

282: \bea

283: \lan \pm | \pm \ran & = & 1 \nonumber \\

284: \lan \pm | \mp \ran & = & 0 . \label{ttwo}

285: \eea

286: %

287: For deflected-down prepared subensemble predictions, vector $ | +

288: \ran_{\theta^\prime}  $ in (\ref{1}) is replaced by $ |  - \ran_{

289: \theta^\prime } =  \cos\frac{ \theta^\prime }{2} | - \ran -  \sin\frac{

290: \theta^\prime }{2} | + \ran  $.

291: 

292: \noindent Spin-up/spin-down vectors, $ | \pm \ran $ , are so-called by

293: analogy with the relation between ordinary classical spin vectors, $ \bs_\pm

294: $, and what would be their SG deflections. However in the

295: two-dimensional Hilbert space H2 which they span, the directions of $ | \pm

296: \ran $  are not antiparallel, but mutually perpendicular, corresponding to

297: dual, mutually exclusive experimental outcomes.

298: %

299: \begin{center}

300: \scalebox{1.5}[1.5]{\includegraphics{fig8mc.ps}}\\

301: figure 7

302: \end{center}

303: %

304: Given projectors $ {\rm P }_{\theta \pm} $ , both vectors are completely

305: specified by relations

306: %

307: \bea

308: {\rm P }_{\theta \pm } | \pm \ran_\theta & = & (1) \cdot | \pm \ran_\theta =

309: | \pm \ran_\theta  \nonumber \\

310: {\rm P }_{\theta \pm } | \mp \ran_\theta & = & (0) \cdot | \mp \ran_\theta =

311: 0

312: \label{2}

313: \eea

314: %

315: called eigenvalue (proper- value) relations, where (\ref{2}) states that $ |

316: + \ran_\theta $ is an eigenvector of both projector operators $ {\rm P

317: }_{\theta + }$ and $ {\rm P }_{\theta - }$  having eigenvalues 1 and 0,

318: respectively; these are sometimes called "yes" and "no" eigenvalues, and

319: projectors, accordingly, yes-no operators. In consequence, $ | \pm

320: \ran_\theta

321: $ are also eigenvectors of the quantum mechanical spin operator, $ S_\theta

322: = \frac{1}{2} ( {\rm P }_{\theta + }- {\rm P }_{\theta - })  $ , having

323: eigenvalues $ \pm \frac{1}{2} $

324: %

325: \bea

326: S_\theta | \pm \ran_\theta & = & \pm \frac{1}{2} | \pm \ran_\theta

327: \label{3}

328: \eea

329: %

330: which gives immediately that the vectors $ | \pm \ran_{ \theta^\prime } $

331: with $ \theta^\prime \neq \theta  $  are generally not spin $ S_\theta $

332: eigenvectors.

333: 

334: \noindent Vectors and operators may also be expressed in 2-dimentional

335: matrix form

336: %

337: \begin{equation}

338: | + \ran_\theta =  \left( \begin{array}{r}\cos\frac{\theta }{2} \\

339: \sin\frac{ \theta }{2}  \end{array} \right) \qquad \qquad \qquad \qquad

340: | - \ran_\theta =  \left( \begin{array}{ r }- \sin\frac{\theta }{2} \\

341: \cos\frac{ \theta }{2}  \end{array} \right) \label{2-vecs}

342: \end{equation}

343: %

344: \begin{equation}

345: {\rm P }_{ \theta + } =  \left( \begin{array}{ r r }\cos^{2}\frac{\theta

346: }{2}& \frac{1}{2} \sin\theta \\ \frac{1}{2}\sin\theta & \sin^{2}\frac{

347: \theta }{2}  \end{array} \right) \qquad \qquad

348: {\rm P }_{ \theta - } =  \left( \begin{array}{ r r }\sin^{2}\frac{\theta

349: }{2}& - \frac{1}{2}\sin\theta \\ - \frac{1}{2}\sin\theta & \cos^{2}\frac{

350: \theta }{2}  \end{array} \right)

351: \end{equation}

352: %

353: whereby relations (\ref{1}), (\ref{ttwo}), and (\ref{2}) above

354: are of course preserved. In addition, for each quantization $ \theta $  we

355: have the

356: projector relations

357: %

358: \begin{equation}

359: {\rm P }_{\theta + } + {\rm P }_{\theta - } = \left( \begin{array}{c c} 1 &

360: 0 \\ 0 & 1 \end{array} \right) \equiv { \bf 1 }

361: \end{equation}

362: %

363: %

364: \bea

365: {\rm P }^2_{\theta \pm } & = & {\rm P }_{\theta \pm }

366: \eea

367: %

368: known respectively as completion and projector conditions. Their symbolic

369: evaluation over the ensemble yields the expectation value equation

370: %

371: \bea

372: \lan {\rm P }_{\theta + } \ran_\theta + \lan {\rm P }_{\theta - }

373: \ran_\theta & = & 1

374: \eea

375: %

376: with constraint

377: %

378: \bea

379: \lan {\rm P }_{\theta \pm } \ran_\theta & = & 1 \, {\rm or } \, 0 \, .

380: \eea

381: %

382: It is sometimes of interest to compare sets of eigenvector pairs for

383: selected quantizations. From (\ref{2-vecs}) we tabulate for orientations

384: $\{ \theta \} = \{0,\pi/2, \pi, 3\pi/2 \} $.

385: %

386: \begin{center}

387: \scalebox{0.9}[0.9]{\includegraphics{fig17m.ps}}\\

388: table 1

389: \end{center}

390: %

391: Notice that eigenstates are not necessarily exclusive to single

392: orientations: $ |- \ran_{\pi/2} = |+ \ran_{3 \pi/2} $, so that e.g. from

393: (\ref{nnew}), $ {\rm P }_{\pi/2_-} = {\rm P }_{3 \pi/2 + } $. As the

394: ambiguity is significant to the coming discussion, for emphasis we now make

395: a notational change. We specify a Hilbert space eigenstate now by the

396: spatial direction of its classical counterpart, one related to the other as

397: illustrated in

398: figure (7). Then, $ | \pm \ran_{\theta^\prime } \sim | \bs^\prime_\pm \ran

399: $. $ {\rm P }^\prime_i$  thus projects out the QM state with classical

400: counterpart $ \bs^\prime_i $ measured by an SG apparatus with orientation $

401: \theta^\prime $, which, interchangeably with the corresponding average SG

402: field direction $ \hat{\bB^\prime } $, we shall call the measurement's {\it

403: context}. The two notations are related by

404: %

405: \bea

406: \hat{\bB}^\prime \cdot \hat{\bk} &=& \cos(\theta^\prime ) \nonumber \\

407: \hat{\bs}^\prime_{1,2} & \equiv & \hat{\bs}^\prime_{+,-} = +,-

408: \hat{\bB}^\prime \nonumber \\

409: {\rm P }_{1,2}^\prime &=& {\rm P }_{\theta^\prime +,- }

410: \eea

411: %

412: We therefore have from table 1

413: %

414: \bea

415: {\rm P }_2^{(\pi/2)} &=& {\rm P }_1^{(3\pi/2)}

416: \eea

417: %

418: consequent upon $ \bs_2^{(\pi/2)}= \bs_1^{(3\pi/2)} ( = \hat{\bi }) $: The

419: spin-down SG outcome state for SG orientation $ \theta = \pi/2 $ is the same

420: as the spin-up outcome state for SG orientation $ \theta = 3\pi/2 $   . In

421: the new notation table 1 becomes

422: %

423: \begin{center}

424: \scalebox{0.9}[0.9]{\includegraphics{fig18m.ps}}\\

425: table 2

426: \end{center}

427: %

428: For the purposes of the KS analysis here we will need to consider the

429: structure of measurement outcomes on spin-1 particles, measurements also

430: made by the SG apparatus. The observed mutually exclusive outcomes are:

431: deflection up, deflection down,

432: and no deflection at all.

433: %

434: \begin{center}

435: \scalebox{1.5}[1.5]{\includegraphics{fig5m.ps}}\\

436: figure 8

437: \end{center}

438: %

439: In the now corresponding  {\it three} dimensional Hilbert space of states we

440: have operator

441: relations

442: %

443: \begin{equation}

444: {\rm P }_1  + {\rm P }_2  + {\rm P }_3 = \left( \begin{array}{c c c} 1 & 0 &

445: 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{array} \right) \equiv { \bf 1 }

446: \end{equation}

447: %

448: \bea

449: {\rm P }^2_i  & = & {\rm P }_i

450: \eea

451: %

452: and related expectation-value completion and projection relations

453: %

454: \bea

455: \lan {\rm P }_1 \ran^\prime + \lan {\rm P }_2  \ran^\prime  + \lan {\rm P

456: }_3 \ran^\prime  & = & 1 \nonumber \\

457: \lan {\rm P }_i  \ran^\prime & = & 1 \, \, {\rm or } \, \, 0

458: \qquad \qquad \qquad  {\rm for } \; \bB^\prime = \bB

459: \label{new}

460: \eea

461: %

462: a constrained sum rule (csr) central in the derivation of the KS paradox.

463: Looking ahead, we now tabulate the eigenvector triplets for the four

464: contexts $ \{ \bB \} =

465: \{ \hat{\bk},- \hat{\bi},- \hat{\bk},\hat{\bi} \} $

466: %

467: \begin{center}

468: \scalebox{0.9}[0.9]{\includegraphics{fig19m.ps}}\\

469: table 3

470: \end{center}

471: %

472: Table 3 and csr (\ref{new}) taken together lend themselves to various

473: physical

474: interpretations, most of which may be grouped into one of two main

475: categories: orthodox and realist. The KS contention detailed in section 6 is

476: that an expanded  table 3 (obtained simply by expanding the number of

477: contexts considered) is not amenable to any self-consistant realist

478: interpretation; this is the KS

479: paradox. For this reason, we begin in the next two sections by considering

480: such features of the two broad interpretations as are relevant to the KS

481: discussion.

482: 

483: \section{Orthodox interpretation of SG data and QM

484: description}

485: 

486: \subsection{ Idea of the individual electron state }

487: 

488: There are several ways in which the behavior of individual electrons under

489: the influence of SG measurements is similar to that of quantum-mechanical

490: spin eigenfunctions under the action of projector operators. As a prepared

491: electron deflects either up or down along a measuring SG axis, a projector

492: operator will project an arbitrary state $ \Psi $  into either a spin-up or

493: a spin-down eigenstate characteristic of the SG orientation $ \hat{\bB} $

494: \footnote{ The actual projection on an arbitrary state is by prescription

495: only; a state is said to "collapse" upon measurement to one of the context's

496: eigenstates, though no entirely satisfactory dynamics for the collapse has

497: ever been worked out \cite{125}. The shortcoming is closely related to the

498: so-called measurement problem in QM. }.

499: %

500: \bea

501: {\rm P } | \Psi \ran &  & \stackrel{\longrightarrow }{collapse} \quad |

502: \bs_1 \ran \quad or \quad | \bs_2  \ran

503: \eea

504: %

505: An electron spin prepared along a given measurement axis then will not be

506: altered upon measurement - a deflected up(down) prepared electron deflects

507: up(down) again - just as projector operators project their own eigenstates

508: onto themselves. On the other hand, if the prepared and measurement axis are

509: not identical, the electron deflects up or down with probability amplitude

510: given by the corresponding state projected eigenfunction coefficient

511: %

512: \begin{equation}

513: {\rm P }_{1,2}^\prime | \bs_1 \ran  = \left\{ \begin{array}{l} \cos\frac{

514: \phi }{2}| \bs_1^\prime \ran   \\  \sin\frac{ \phi }{2}| \bs_2^\prime \ran

515: \end{array} \right.  \label{10}

516: \end{equation}

517: %

518: illustrated

519: %

520: \begin{center}

521: \scalebox{1.5}[1.5]{\includegraphics{fig7m.ps}}\\

522: figure 9

523: \end{center}

524: %

525: And so one might well identify the quantum mechanical spin- $ \frac{1}{2} $

526: wave function $|\Psi \ran $ with individual electron states, projector

527: operators with the measuring apparatus, and action of the operator on the

528: wave function with the measuring process, just the

529: identifications in fact made in the conventional or orthodox interpretation

530: of QM \footnote{ See e.g. ref. \cite[pp. \ 108 \& 214]{101} .}. While the

531: assignments are largely unimportant to the practical workings of the

532: formalism, the calculation of expectation values, they offer a picture of

533: the world of individual electrons, a kind of metaphysics.

534: 

535: \noindent One difficulty with the interpretation is that spin-vectors are

536: particular to their quantization, having definite projections only along the

537: quantized directions of their measurement axis; in all other directions the

538: projection is not well-defined. The same then in this interpretation for the

539: electron spin itself; the orthodox electron does not possess a

540: spin-projection along a given direction except upon measurement (and prior

541: to subsequent distinct measurements); unlike ordinary classical spin

542: projections, an electron's seems to

543: exist only along one set of directions at a time.

544: 

545: \noindent Another difficulty is that "identical" electrons - those of a

546: given ensemble and so represented by the same spin function - generally

547: exhibit different behaviors when subjected to the same conditions: Some

548: electrons of a deflected-up subensemble e.g. will deflect up, others down,

549: when their spin projections are measured along a direction different from

550: the preparation axis. The phenomena is at odds with the principle of the

551: identity of indiscernibles \cite{122} (see also discussion in \cite[p. \

552: 8]{116}, and "sufficient reason" principle in \cite[p. \ 266]{104}  ).

553: 

554: \noindent Finally, the probabilities of QM have meaning to the

555: experimentalist only in reference to ensemble measurements. \footnote{ see

556: ref. \cite[p. \ 2]{102} and \cite[p. \ 36]{103} }

557: 

558: \subsection{ Orthodox interpretation of csr (\ref{new})  }

559: 

560: The value of the expectation  $ \lan  {\rm P }_i  \ran_\Psi  $ answers the

561: question whether a particle in state $ \Psi $ has a definite spin in context

562: $ \bB $ : 1 or 0 for yes, no otherwise. From (\ref{new}) we see that the

563: answer is "yes" generally when the particle has been prepared within the

564: measurement context of interest, $ |\Psi \ran \to |\bs_j^\prime \ran $ with

565: $ \bB^\prime = \bB $. But the answer might also be yes for a different

566: preparation context,  $ \bB^\prime \neq \bB $, provided the classical

567: companions to Hilbert space vectors $ | \bs_j^\prime \ran  $ and $ | \bs_i

568: \ran  $, $ \bs_j^\prime \, \& \, \bs_i $, coincide for some i and j. We

569: illustrate such an instance

570: %

571: \begin{center}

572: \scalebox{1.5}[1.5]{\includegraphics{fig4m.ps}}\\

573: figure 10

574: \end{center}

575: %

576: In this case the projection measurement with result corresponding to that

577: for classical spin direction $ \bs_2 = \bs_2^\prime $ can be made within

578: either context $ \bB $ or $ \bB^\prime $ with equal results, and is

579: therefore said to be context independent (wrt contexts $ \bB $ and $

580: \bB^\prime $ ), or noncontextual. A prepared particle $ | \bs_j^\prime \ran

581: $

582: might therefore have a definite spin in all contexts with quantization $

583: \bs_j^\prime $  and can be said to {\it possess} the corresponding set of

584: projector operator eigenvalues. In all other contexts its spin projections

585: are not definite, and the particle could not be said to possess the

586: corresponding projector eigenvalues.

587: 

588: It is easy to see that the number of noncontextuality constraints imposed by

589: this interpretation on a set of sum rules (\ref{new}) corresponding to a

590: large number of contexts considered simultaneously would be relatively

591: small. Not so in the realist interpretation, as we shall see in the next

592: section. It is this consequence of the realist interpretation that the KS

593: analysis in the section following exploits to draw a self-contradiction.

594: 

595: \section{Realist interpretation of SG data and QM description }

596: 

597: \subsection{ Deterministically-held possessed values and the realist

598: extension to csr (\ref{new}) }

599: 

600: In the realist view there is a sense in which an electron has definite spin

601: projections in all directions at all time; its physical

602: characteristics combined with preparation determines {\it  unambiguously}

603: whether an

604: electron will deflect up or down in a given Stern-Gerlach measurement,

605: nothing left to chance. "The good lord does not throw dice", is how Einstein

606: is reported to have put it \cite[p. \ 190]{104}. Then, in the event that

607: otherwise identical electrons have different deflections their preparations,

608: their causal histories, must also have been different.

609: 

610: \noindent { \it In this sense } it might be said that an electron is in

611: possession of a deflection value, "up" or "down", and hence in possession of

612: its spin projection, prior to measurement. In much the same sense we say

613: that a massive object possess a weight, referring to the weight value $

614: \nu(w) $ a scale will read when the object is placed on it.This does not

615: conflict with the possibility that one scale gives a different reading from

616: another; one may speak of a body's terrestrial weight, its weight on the

617: moon,

618: on mars, etc.. A massive body, like the realist electron, is at all times in

619: possession of an infinite number of measurement values, the set of values

620: that (for the body) weight scales may measure for each of the infinitely

621: many possible measurement situations. These values are deterministically

622: well-defined and so, given the object's physical characteristics and the

623: deterministic factors acting upon it, the causal chain leading up to the

624: measurements, may be identified with that oject. By this identification - in

625: this sense - we say that the

626: values are possessed by the object - weights of a body, spin projections

627: of an electron.

628: 

629: \noindent A body's weight values might of course be determined from its

630: mass, given the relevant celestial masses and the law of gravity; in a world

631: of weight measurements however a statement of either is equivalent.

632: Likewise, in absence of a knowledge of some underlying electron spin and the

633: dynamics of microscopic spin-measurements, we may meaningfully identify a

634: set of possible measurement results $ \{ \nu(s) \} $ with an electron.

635: Fundamental to the realist view is that an object's possessed values are, in

636: principal, verifiable measurement values.

637: 

638: \noindent The description is compatible with the realist tenet that

639: %

640: \begin{equation}

641: {\it Distinct \,\, phenomena \,\, supervene \,\, upon \,\, distinct \,\,

642: causes \,\,

643: only } \footnote{ cf. opposing accounts of the "realist view" in \cite[p. \

644: 19]{116} and \cite[pp. \ 12-13]{118} }. \label{cause}

645: \end{equation}

646: %

647: on which is based the EPR reality criterion

648: %

649: \begin{quotation}

650: \small

651: \noindent If, without in any way disturbing a system, we can predict with

652: certainty (i.e., with probability equal to unity) the value of a physical

653: quantity, then there exists an element of physical reality to this physical

654: quantity.

655: \end{quotation}

656: %

657: which set the stage for their famous argument in defense of realism

658: \cite{105} and lead eventually to the Bell inequality constraint on the

659: same \cite{5}. In Bell's analysis, the above description of electron

660: possessed values $ \{ \nu(s) \} $ also appear, depending too upon the manner

661: in which the spin measurements are taken, their contexts. And it is only

662: upon stringent space-time constraints that context dependence is ruled out.

663: 

664: \noindent We now use the notation $ \{ \nu(s) \}^\prime $ to denote the set

665: of electron spin projection values measurable in context $ \bB $, though

666: possessed by the particle while in context $ \bB^\prime $ . Similarly for

667: the set of projector (yes-no) values

668: %

669: \bea

670: \{ \nu({\rm P } ) \}^\prime & \equiv &  \{ \nu^\prime({\rm P }_1 ),

671: \nu^\prime({\rm P }_2  ), ...  \nu^\prime({\rm P }_n )  \}

672: \eea

673: %

674: where $ \nu^\prime({\rm P }_i ) $ is the yes-no value for the $ i^{th} $

675: quantized direction (corresponding projector eigenvalues in descending

676: order) of context $ \bB $, possessed by a particle while in context $

677: \bB^\prime $. We identify realist possessed values with QM operators in this

678: notation only for the sake of comparison with QM expectation values with

679: which they sometime coincide; the value $ \nu({\rm P } ) $ simply answers

680: the question whether or not a spin-measured quantum particle would scatter

681: in the same direction as a classical particle of spin  $ \bs $, and is not

682: to be understood as a formal mapping of the QM projector  $ {\rm P } $. The

683: correlation between expected and experimental values is of primary

684: importance in the realist view, that between expected values and the QM

685: formalism (or any other) of secondary importance; the extent to which

686: eigenvalues of a QM operator are observed values is the same to which the

687: realist possessed value $ \nu(\bO) $ is an eigenvalue

688: of the QM operator $ \bO $.

689: 

690: \noindent By realist extension of csr (\ref{new}) to all contexts, then, we

691: have:

692: %

693: \bea

694: \nu^\prime({\rm P }_1 )+\nu^\prime({\rm P }_2 )+\nu^\prime({\rm P }_3 ) & =

695: & 1  \nonumber \\

696:   \nu^\prime({\rm P }_j  ) & = & 1 \, or \, 0 \qquad \qquad \qquad  \forall

697: \quad \bB^\prime . \label{real eqns}

698: \eea

699: %

700: Lastly, we may attribute to a  hypothetical realist state function, one

701: completely describing the state of an individual particle, the complete set

702: of the particle's possessed values

703: %

704: \bea

705: \psi &=& \psi [ \{ \{ \nu({\rm P } ) \}^\prime \} ]  \qquad \qquad  \forall

706: \quad \bB , \; \bB^\prime. \label{g}

707: \eea

708: %

709: The realist particle thus represented by $ \psi $ has definite spin

710: projections in all H-spatial directions, simultaneously in all contexts.

711: 

712: \subsection{Relation between realist $ \psi $ and QM $ \Psi $  }

713: 

714: \noindent We symbolically represent the value assignment for an

715: electron-spin measurement

716: %

717: \bea

718: \lan \psi | \bS | \psi \ran  &=& \nu(\bs )  \to \pm \frac{1}{2} \, .

719: \label{c}

720: \eea

721: %

722: For sets of realist possessed values of mutually non-commuting contexts,

723: however,

724: %

725: \bea

726: \left[  \bS_\theta , \bS_{ \theta^{ \prime } } \right]  & \neq & 0  \, ,

727: \eea

728: %

729: QM predicts and observation confirms that a measurement of one set alters

730: the elements of the other; determination of such simultaneous possessed 

731: value sets is therefore not possible.

732: 

733: \noindent We illustrate a possible difficulty to realist thinking that this

734: resolves by first reverting back briefly to the former notation and consider

735: the QM operational identity

736: %

737: \bea

738: \bS_{\pi/4} &=& (\bS_0 + \bS_{\pi/2} )/\surd2 \label{d1}

739: \eea

740: %

741: with the above realist value assignments (\ref{real eqns}) applied to its QM

742: expectation

743: %

744: \bea

745: \lan \Psi | \bS_{\pi/4 } | \Psi \ran &=&  ( \lan \Psi | \bS_{0} | \Psi \ran

746: + \lan \Psi | \bS_{\pi/2 } | \Psi \ran )/\surd2  \rightarrow \pm \frac{1}{2}

747: = (\pm \frac{1}{2} \pm \frac{1}{2})/\surd2 \label{h0}

748: \eea

749: %

750: a statement manifestly false for all $\pm $ combinations. The mistake

751: results from an essential difference between the eigenvalues $\nu(\bO) $ and

752: expectation values $ \lan \Psi | \bO | \Psi \ran $ of an operator $ \bO $.

753: We refer however to the use of relation (\ref{d1}) for its {\it meaning } 

754: which lies in the agreement obtained between its

755: expectation value and the experimental average of measurements made upon

756: many, many electrons for each of the three indicated contexts, i.e. in an 

757: agreement

758: with empirical relations between sub-ensemble averages. In practice, then,

759: (\ref{d1}) pertains to three entirely disjoint electron sub-ensembles, each

760: measured within one of the three contexts, $ \{ 0, \pi/4, \pi/2 \}  $, so

761: that no single electron spin is measured more than once; the meaning of

762: (\ref{d1}) derives from the identification of quantum mechanical state

763: functions $ \Psi $ with subensembles \cite{111}, and not from a direct

764: association with the possessed values of individual ensemble members

765: described by the realist $ \psi $. With $ | \Psi \ran $ as the QM state

766: function, the expectation value relation

767: %

768: \bea

769: \lan \Psi | \bS_{\pi/4 } | \Psi \ran &=&  ( \lan \Psi | \bS_{0} | \Psi \ran

770: + \lan \Psi | \bS_{\pi/2 } | \Psi \ran )/\surd2     \label{h1}

771: \eea

772: %

773: is found to correspond to the observed relation

774: %

775: \bea

776: \bar{ \bS}_{\pi/4} &=& ( \bar{ \bS}_{0} + \bar{ \bS}_{\pi/2}) /\surd2

777: \label{h2}

778: \eea

779: %

780: where

781: %

782: \bea

783: \bar{\bS}_\theta & \equiv & \frac{1}{2}( n_+ - n_- )/N

784: \eea

785: %

786: $ n_\pm $  the number of ensemble electrons observed to deflect up/down

787: in context $ \theta $, and $ N = n_+ +  n_- \rightarrow \infty $,  the total

788: number of electrons. Although the QM formalism does not itself attach to its

789: state functions a physical interpretation, in use $ \Psi $ is clearly to be

790: associated with sub-ensembles, in line with its realist interpretation

791: \footnote{ For an alternative and unconventional view see \cite{124}. With

792: regard to the special-case arguments put forward there, one might also

793: consider the special case of homogeneous ensembles.}.  An expansion of the

794: QM $\Psi $ over hypothetical realist electron states $ \psi $ might take the

795: form

796: %

797: \bea

798: |\Psi \ran &=& ( n_+  |+\ran_\theta +n_- |-\ran_\theta )/N \nonumber \\

799: &=& ( n_1  |\bs_1 \ran +n_2 |\bs_2 \ran )/N \label{Psi}

800: \eea

801: %

802: with

803: %

804: \bea

805: | \bs_i \ran &=& \sum_{ \{ \nu({\rm P }_i  ) = 1 \}} | \psi [ \{ \nu({\rm P

806: }^\prime ) \}^{\prime \prime} ] \ran \label{h3}

807: \eea

808: %

809: normalized to 1. Such an expansion leads immediately from the QM prediction

810: (\ref{h1}) to the observed (\ref{h2}).  The terms averaged over, $ \{\nu

811: ({\rm P }^\prime ) \}^{\prime \prime} $, with their contextual dependence,

812: may be understood as the sub-ensemble's $ \bB $-context hidden variables.

813: 

814: \noindent As $ \hat{\bB} $ context eigenstates $ | \bs_i \ran $ will

815: generally not have definite values in other contexts, we have

816: %

817: \bea

818: 0 \leq & \lan \bs_j^\prime | {\rm P }_i  | \bs_j^\prime \ran  &  \leq 1

819: \eea

820: %

821: although for realist possessed values

822: %

823: \bea

824: \nu^\prime ({\rm P }_i ) &=& 1\, {\rm or } \, 0 \, \qquad \forall \quad

825: \bB^\prime.    \label{h5}

826: \eea

827: %

828: A QM state function $ |\Psi \ran $ of (\ref{Psi} ) on the other hand has

829: definite values in at most one context and is therefore said in the realist

830: view to give a physically incomplete picture of the individual electron.

831: This difference forms the basis of our reconsideration of the KS paradox in

832: section 7.

833: 

834: \section{Kochen-Specker paradox }

835: 

836: \subsection{derivation of the paradox}

837: 

838: In line with the discussion in section 5 the following might be proposed as

839: a realist proposition

840: %

841: \begin{quotation}

842: \noindent p(1):  All real state quantities (spin, mass, etc.) have definite

843: values at all times.

844: \end{quotation}

845: %

846: which is known as the principle of value-definiteness. Pairing with this a 

847: second

848: proposition

849: %

850: \begin{quotation}

851: \noindent p(2):  The value of a real quantity does not depend on how it is

852: measured

853: \end{quotation}

854: %

855: called noncontextuality, and applying them together to the csr's (\ref{real 

856: eqns})

857: draws a contradiction in any Hilbert space of greater than two dimensions. 

858: As

859: the Hilbert space dimension for a particle of spin "s" is given by 2s+1, KS

860: shows that it is not possible to apply both p(1) and p(2) consistently for a

861: particle of spin $ s \geq 1 $; this is the KS

862: paradox \cite{106, 107}.

863: 

864: \noindent Let us consider the simultaneous application of the propositions

865: on (\ref{real eqns}).

866: From p(1) all equations and constraints hold simultaneously at this time

867: %

868: \begin{enumerate}

869: \item $ \nu^\prime ({\rm P }) = 0 \, {\rm or } \, 1 $

870: \item  for all $ \bB^\prime $

871: \end{enumerate}

872: %

873: and from p(2) the values are independent of context:

874: %

875: \begin{enumerate}

876: \item $  \nu^\prime ({\rm P } ) =   \nu^{\prime \prime  }({\rm P } ) \to

877: \nu ({\rm P } ) $.

878: \end{enumerate}

879: %

880: In their original work the authors next proceed with a lengthy formal 

881: argument, a simplification of which is the following :

882: 

883: \noindent The numbers $  \nu ({\rm P } ) $ are the eigenvalues of  Hilbert 

884: space

885: vectors $ | \bs_j \ran  $. According to the realist csr's (\ref{real eqns})

886: it must then be possible to assign to each mutually orthogonal set of three

887: vectors the values 0, 0, and 1. As the Hilbert space of QM state vectors,

888: H3, is complex, a violation of this rule in a Real space of the same

889: dimension will suffice for the negative proof of interest (R3 being a

890: subset of H3). Note however that the corresponding real vectors in the

891: mapping $ QM: R3 \to H3 \supseteq R3 $ (analogous to the 2-dimentional case

892: pictured in figure 7) do not coincide.

893: 

894: \noindent To begin, KS show in the following

895: way that two vectors separated by an angle

896: %

897: \bea

898: \phi  &\leq & \cos^{-1}(\surd8/3) \label{angle}

899: \eea

900: %

901: must have the same eigenvalue assignment, 0 or 1. Consider the three sets of

902: mutually orthogonal triplets

903: %

904: \bea

905: \{ |\bs_1^{(1)}>, |\bs_2^{(1)}>, |\bs_3^{(1)}> \} , & \{ |\bs_1^{(2)}>,

906: |\bs_2^{(2)}>, |\bs_3^{(2)}> \} , & \{ |\bs_1^{(3)}>, |\bs_2^{(3)}>,

907: |\bs_3^{(3)}> \} \label{nnnnew}

908: \eea

909: %

910: in R3 together with a tenth, $ |\bs_0> $, defined as

911: %

912: \bea

913: |\bs_0 \quad > & =& -x y \bi +x \bj -\bk   \nonumber \\

914: |\bs_1^{(1)}> &=& \bj+x \bk \nonumber \\

915: |\bs_2^{(1)}> & =& \bi    \nonumber \\

916: |\bs_3^{(1)}> & =& -x \bj +\bk   \nonumber \\

917: |\bs_1^{(2)}> & =& y \bi - \bj   \nonumber \\

918: |\bs_2^{(2)}> & =&  \bi + y \bj   \nonumber \\

919: |\bs_3^{(2)}> & =& \bk   \nonumber \\

920: |\bs_1^{(3)}> & =& - \bi -y \bj -x y\bk   \nonumber \\

921: |\bs_2^{(3)}> & =&   - y (1+x^2) \bi +(1-x^2 y^2 ) \bj + x (1+y^2) \bk

922: \nonumber \\

923: |\bs_3^{(3)}> & =&   -x y^3 (1+x^2) \bi + x (1+ 2 y^2 +x^2 y^2 ) \bj -

924: (1+y^2 )  \bk \label{mnew}

925: \eea

926: %

927: for arbitrary numbers  x  \&  y. Now the angle $ \phi $ between $ |\bs_0> $

928: and $ |\bs_3^{(3)}> $  found from their scalar product is given by relation

929: %

930: \bea

931: \cos(\phi ) &=& [1+x^2+y^2+x^2 y^2 (2+x^2+y^2+x^2 y^2) ] / ( ||\bs_0>||

932: ||\bs_3^{(3)}>|| )

933: \eea

934: %

935: whose rhs minimization is directly found at $ \surd8/3 $, obtained for  $ x 

936: = y = \pm 1 $.

937: Therefore, $ 0 \leq \phi \leq \cos^{-1}(\surd8/3) $. In addition to the

938: explicit vector triplet constraints (\ref{real eqns}), by the same

939: consequence, no perpendicular pair of vectors may be assigned number "1",

940: since taken together with a mutually perpendicular third they would form a

941: triplet. One finds then from the additional relations

942: %

943: \bea

944: |\bs_0 \quad > & \perp & |\bs_1^{(1)}>, |\bs_2^{(2)}>, {\rm and } 

945: |\bs_2^{(3)}>

946: \nonumber \\

947: |\bs_1^{(3)}> & \perp & |\bs_3^{(1)}> { \rm and } |\bs_1^{(2)}> \nonumber \\

948: |\bs_2^{(1)}> & \perp & |\bs_3^{(2)}> \label{nnnew}

949: \eea

950: %

951: that the only possible assignments for the set of ten are those for which

952: $|\bs_0> $ and $ |\bs_3^{(3)}> $ have the same eigenvalue, $ \nu({\rm P }_0)

953: = \nu({\rm P }_3^{(3)}) = 0 \, {\rm or } \, 1 $, demonstrated by simply 

954: trying out each of the other two possible assignments while respecting constraints 

955: implicit in

956: (\ref{nnnnew}) and (\ref{nnnew}). As only the relative  $|\bs_0> $ to $

957: |\bs_3^{(3)}> $ directions are relevant, the result establishes for

958: KS that in realist thinking any two vectors in R3 of angular separation $ 0

959: \leq \phi \leq \cos^{-1}(\surd8/3) $ must have the same eigenvalue

960: assignment, 0 or 1.

961: 

962: \noindent We now take the full set $ \{|\bs_i^{(j)}> \}$ of (\ref{mnew}) for

963: $ \phi = 18^o ( < \cos^{-1}(\surd8/3) )  $, and for convenience perform a

964: rotation on the coordinate system such that $ |\bs_2^{(3)}> $ points along

965: the positive y-axis and $ |\bs_0> $ along the positive z-axis. We show here

966: three key vectors of the set scaled to 100, and table triplets according to

967: context $\hat{\bB}_i $.

968: %

969: \begin{center}

970: \scalebox{1.5}[1.5]{\includegraphics{fig13m.ps}}\\

971: figure 11

972: \end{center}

973: %

974: From these we generate four new sets of nine vectors by rotating the

975: original about $ |\bs_2^{(3)}> $ in four $ 18^o $ increments, constraining 

976: successive

977: $|\bs_3^{(3)}> $'s thereby to the same value assignment, and

978: generate five additional sets by first rotating the last set by $ 90^o $

979: about its $ |\bs_3^{(15)}> $, then rotating in five more $ 18^o $ increments

980: about the rotated $ |\bs_2^{(15)}> $ . For a final five we follow the same

981: procedure to complete the first-quadrant sweep of $ |\bs_3 > $:  $

982: |\bs_3^{(15)}> ( = |\bs_2^{(33)}> ) \to |\bs_3^{(30)}> ( = |\bs_2^{(3)}> )

983: \to |\bs_3^{(45)}> ( = |\bs_2^{(18)}> ) $ ,  on the $ \{

984: \hat{\bi},\hat{\bj},\hat{\bk} \} $ triad

985: %

986: \begin{center}

987: \scalebox{1.5}[1.5]{\includegraphics{fig12m.ps}}\\

988: figure 12

989: \end{center}

990: %

991: each element of which, again, constrained to the same value assignment. But

992: this immediately violates csr's (\ref{real eqns}) as $ \{

993: \hat{\bi},\hat{\bj},\hat{\bk} \} $ is itself a mutually perpendicular

994: triplet. The violation is taken to demonstrate an inconsistency in the

995: realist conception of reality as given by propositions p(1) and p(2) and 

996: thus concludes

997: the KS paradox.

998: 

999: \noindent This result may also be illustrated visually. For the tabular form 

1000: \cite{107} consider first the vectors involved in the 15

1001: orientations of the set of 9 as represented in figure (12); we list here the

1002: total 117 vectors (  = $ 15 \times 9 - 15 - 3$, accounting also for the 15 $

1003: |\bs_2> $ and 3 $ |\bs_3>$ overlaps ), sub-grouping as in figure (11)

1004: according to context

1005: %

1006: \begin{center}

1007: \scalebox{0.9}[0.9]{\includegraphics{fig20ma.ps}}\\

1008: \end{center}

1009: %

1010: \begin{center}

1011: \scalebox{0.9}[0.9]{\includegraphics{fig20mb.ps}}\\

1012: \end{center}

1013: %

1014: \begin{center}

1015: \scalebox{0.9}[0.9]{\includegraphics{fig20mc.ps}}\\

1016: \end{center}

1017: %

1018: \begin{center}

1019: \scalebox{0.9}[0.9]{\includegraphics{fig20md.ps}}\\

1020: \end{center}

1021: %

1022: \begin{center}

1023: \scalebox{0.9}[0.9]{\includegraphics{fig20me.ps}}\\

1024: table 4

1025: \end{center}

1026: %

1027: Then from the following rules

1028: %

1029: \begin{enumerate}

1030: \item The eigenvector whose projector has value 1(0) is assigned the color

1031: white (black).

1032: \item Two orthogonal vectors may not both be colored white.

1033: \item The assignments are independent of context.

1034: \end{enumerate}

1035: %

1036: we attempt to color the vector boxes of table 4. That the coloring

1037: assignments cannot be made consistently in accordance with the rules

1038: coincides with the paradox.

1039: 

1040: \noindent Finally, for a two-dimensional graphical form of the paradox 

1041: \cite{108} we represent vectors

1042: by small open circles, and orthogonality between pairs of vectors by a

1043: connecting straight line;  a triangle with open circles at the vertices then

1044: represents a mutually orthogonal triplet, and our beginning set of vectors

1045: described in figure (11) has the KS diagram representation

1046: %

1047: \begin{center}

1048: \scalebox{0.9}[0.9]{\includegraphics{fig14m.ps}}\\

1049: figure 13

1050: \end{center}

1051: %

1052: Here the constraint that $|\bs_0> $ and $ |\bs_3^{(3)}> $ have the same

1053: eigenvalue assignment for example follows from the inability to assign 

1054: different colors

1055: to their circled vertices in accordance with the aforementioned coloring

1056: scheme for the corresponding vector boxes. If we then stretch the diagram a

1057: little to one side

1058: %

1059: \begin{center}

1060: \scalebox{0.9}[0.9]{\includegraphics{fig15m.ps}}\\

1061: figure 14

1062: \end{center}

1063: %

1064: and perform the 14 rotations about the appropriate $ |\bs_2> $ points (along 

1065: with the two rotations about $ |\bs_3> $ points) as represented in

1066: figure 12, we generate the full KS diagram

1067: %

1068: \begin{center}

1069: \scalebox{0.9}[0.9]{\includegraphics{fig16m.ps}}\\

1070: figure 15

1071: \end{center}

1072: %

1073: said to resemble a cat's cradle.

1074: Here too, the inability to faithfully apply the coloring scheme to

1075: individual circled vertices is understood as a manifestation of inconsistency in tealist thinking on microscopic measurement. The difficulties presented by these two 

1076: visualizations

1077: are known as the KS {\it coloring problem}.

1078: 

1079: \subsection{Von Neumann's paradox and resolution}

1080: 

1081: In the Summer of 1932 eminent physicist and mathematician J. von Neumann

1082: devised a proof (VN) against the possibility of a consistent realist

1083: interpretation of QM \cite{109}, a precursor of the KS paradox, by way of the operator avarage relation

1084: %

1085: \bea

1086: \bA = \bB + \bC  & \Rightarrow & \lan \Psi | \bA|\Psi \ran = \lan \Psi |

1087: \bB|\Psi \ran + \lan \Psi | \bC | \Psi \ran

1088: \eea

1089: %

1090: similar to our equation (\ref{h0}) above.

1091: Operators $\bB$ and $\bC $ here do not commute, and the rhs terms are

1092: assumed to take on realist possessed values \cite[p. \ 268]{104}. This last

1093: condition is the VN additivity postulate P.IV from which derives the

1094: expectation value formula

1095: %

1096: \bea

1097: \lan O \ran &=& {\rm Tr } (W O) \label{trace}

1098: \eea

1099: %

1100: where Tr is the trace operator and W a statistical operator independent of

1101: observable O and descriptive of the system under observation. For the hypothetical value-definite

1102: states $ \psi $, so-called "dispersion -free" states   

1103: %

1104: \bea

1105: \lan \psi | O^2 | \psi \ran - \lan \psi | O | \psi\ran^2 &=& 0 \label{disp}

1106: \eea

1107: %

1108: in the language of the time, the formula

1109: proves problematic in the following way: The statistical operator describing the

1110: realist system $ \psi $ is just the projector $ W_\psi = | \psi \ran \lan

1111: \psi | $. Choosing for O then an arbitrary projector $ |\phi \ran \lan \phi

1112: |  $ leads from (\ref{trace}) and (\ref{disp}) to the statement

1113: %

1114: \bea

1115: \lan \phi | {\rm W }_\psi  | \phi \ran &=& 0 \, {\rm or } \, 1  \label{stat}

1116: \eea

1117: %

1118: for all possible QM states $\phi $ which cannot be true as the left-hand

1119: side is continuous in $ \phi $ and the right hand side discontinuous. This cast doubts on the very existence of dispersion-free

1120: states $ {\rm W }_\psi $ \footnote{ I don't know that its been pointed out

1121: in the literature \cite{114} that the above lhs continuity is

1122: consequent upon a separate application of additivity by which two

1123: dispersion-free states of opposite eigenvalues, $ \psi_1 \, {\rm and } \,

1124: \psi_0 $, are joined by a connecting vector $ \Psi(x) = a_x \psi_0 + b_x

1125: \psi_1 $, for properly chosen functions $ a_x $ and $ b_x $ ( See e.g.

1126: reference \cite[p. \ 294]{104} ). Such an operator,  $ O =  |\Psi \ran \lan

1127: \Psi |  $ is not an observable in the Hilbert space of dispersion-free

1128: states under consideration. In light of this, the von Neumann proof might be

1129: understood to follow from an assumption, under additivity, of the existence

1130: of dispersive states in, under realism, the Hilbert space of dispersive-free

1131: states: $ \lan \psi | \Psi \ran = 0 $ or $ 1 $ for all $ \Psi $. }.

1132: 

1133: \noindent This VN impossibility proof has long been understood to follow from a

1134: misapplication of its additivity postulate \footnote{ See J. Bell,

1135: ref.\cite{111}. A more thorough exposition is given in \cite[pp. \

1136: 25-34]{116} } which is experimentally valid for expectation values, though, as shown

1137: in section 5.2, not applicable to real-possessed values. The violation follows from simple statistics:

1138: ensemble values ($ \sim \Psi^2  $ ) are generally not identical with

1139: ensemble-member values ( $\sim \psi^2 $).

1140: The Von Neumann paradox rests upon an illicit application of proposition

1141: p(1) to quantum mechanical sub-ensembles via equations of QM which $\Psi $

1142: alone satisfies:

1143: %

1144: \begin{center}

1145: $\Psi $ + p(1) $ \Rightarrow $ VN Paradox

1146: \end{center}

1147: %

1148: ( $\Psi $ implying p(2) ). It has the schematic resolution

1149: %

1150: \begin{center}

1151: $ \Psi  \rightarrow  \psi $

1152: \end{center}

1153: %

1154: or equally well,

1155: %

1156: \begin{center}

1157: p(1) $ \rightarrow $ p(2).

1158: \end{center}

1159: %

1160: We have then

1161: %

1162: \begin{equation}

1163: \left. \begin{array}{c} \psi  + p(1) \\ {\rm or } \\ \Psi + p(2) \end{array}

1164:   \right\} \Rightarrow  \qquad {\rm no \; VN \; Paradox }

1165: \end{equation}

1166: %

1167: \subsection{Gleason's result }

1168: 

1169: In response to the objectionable additivity postulate necessary to the VN

1170: proof, A. M. Gleason derived the main there formula (\ref{trace}) by means of an

1171: acceptable additivity for commuting operators \cite{113}, as appears e.g. in

1172: the above csr's (\ref{real eqns}), in addition to a tacit assumption of

1173: noncontextuality first noticed only years later by Bell \cite{111}. By

1174: demonstrating via (\ref{trace}) the necessary continuity of mappings from H3

1175: and higher dimensions to R1, i.e. that Hilbert space projection operators ${\rm P

1176: }_i $ are mapped continuously onto the closed interval [1,0],  Gleason

1177: proves under noncontextuality the impossibility of the realist necessarily

1178: discontinuous mapping $\nu : {\rm P } \to R1 , \,\, \nu( {\rm P }) = 0 \,

1179: {\rm or }\, 1 $ , and so the non-existence of $\psi $ and an inconsistency

1180: in realist hidden variable theories. The key result of Kochen and Specker,

1181: constraint (\ref{angle}) , then gives an upper-bound estimate on the $

1182: \nu^\prime(\phi) $ implicit to noncontextual value-definite hidden variable

1183: theories.

1184: 

1185: \section{the KS paradox reconsidered }

1186: 

1187: The KS proof improves on the VN proof in that it in effect applies the

1188: constraints of value definiteness, proposition p(1), appropriately to the

1189: hypothetical realist state function $ \psi $ ($ \sim \nu $ ). Proposition

1190: p(2), noncontextuality, on the other hand has largely been inferred

1191: indirectly as an extension to realist thinking. For example, from Einstein's rhetorical

1192: %

1193: \begin{quotation}

1194: Do you believe the moon is really there if no one is looking? \cite{119} (e)

1195: \end{quotation}

1196: %

1197: which suggests an independence of reality from its observation.

1198: 

1199: \noindent Let us briefly take a closer look. As the statement addresses the question of existence itself, the observation-independence or

1200: noncontextuality inferred from it is ontological\footnote{ Ontological in the sense that it concerns the principles

1201: and causes of being.... Unfortunately for the writer, the term

1202: "ontological contextualism" is already in use \cite{117} though with a meaning

1203: it seems closer to that of causal contextuality. I use it here sparingly

1204: in the first sense. }: There is an objective reality "out there"

1205: with an {\it existence } independent of observation \footnote{ See e.g. K.

1206: Popper in ref.\cite[p. \ 2]{112}  }. The aforementioned reality criterion of

1207: section 5.1, however concerns an observation

1208: dependence of a different, more mundane sort: The observation-dependence of

1209: states of existence, of {\it states } (which addresses e.g. the question whether

1210: an electron's spin is affected by an observation of it ), not of the very

1211: existence of the thing observed. The clear indication is that,

1212: notwithstanding the select elements that meet the reality criterion - a

1213: sufficient criterion - the value of a physical quantity does indeed depend

1214: upon how it is obtained, upon the manner in which the object under

1215: observation is affected, disturbed during measurement. Such a contextual

1216: dependence is causal and the values thus measured causally contextual - precisely the sort of effect that experimentalists in EPR tests go to great lengths to guard against \cite{110}. As J. Bell writes,

1217: "The result of an observation may reasonably depend not only on the state of

1218: the system... but also on the complete  disposition of the [measuring]

1219: apparatus." \cite{111}. But these categories are confused in the KS and Gleason

1220: analysis, realist dismissal of ontological contextuality misinterpreted as a

1221: dismissal of causal contextuality, leading, finally, to mistaking p(2) as a realist

1222: proposition.

1223: 

1224: \noindent While the von Neumann proof relies on an explicit misapplication

1225: of realist value-definiteness within the QM formalism, as shown in 5.1, in its

1226: misapplication of additivity via the equations of QM such as (\ref{d1})

1227: to realist states $ \psi $, it is also seen to violate causal contextuality; in the single QM expectation-value equation

1228: (\ref{h1}) the relevant contextual effects are washed out in the average. KS

1229: and Gleason achieve the same ends by means of properly single

1230: context equations, though several taken simultaneously in terms of the csr's of (\ref{real

1231: eqns}) for mutually exclusive contexts. And so in principle both commit the same violation in unduly

1232: constraining the most general causality. Again, to illustrate, there is

1233: nothing in realist thinking on the spin-1 measurements

1234: %

1235: \begin{center}

1236: \scalebox{1.5}[1.5]{\includegraphics{fig9m.ps}}\\

1237: figure 16

1238: \end{center}

1239: %

1240: ($\theta - \theta^\prime = 90^o $) to suggest the counterfactual implied in

1241: the noncontextual constraint $\nu (P_1) = \nu^\prime (P_2) $ ( or, with

1242: reference to figure 10, the constraint $\nu (P_2) = \nu^\prime (P_2) $ ).

1243: 

1244: \noindent The KS diagram of figure (15) describes a set of experiments each

1245: of which may in principle be performed; accordingly, the diagram itself may

1246: always be constructed. Its coloring however depends somewhat upon interpretation of QM wavefunction.

1247: While both orthodox and realist interpretations achieve coloring consistency

1248: by means of various shadings, indeed in violation of the coloring rules,

1249: their mechanisms differ: Indeterminism on one hand, causal contextuality on

1250: the other. At issue here is the KS contention that while orthodox coloring

1251: is constrained only by the rules governing individual triplets, those that

1252: might be derived from the csr (\ref{new}), realist coloring is constrained

1253: by the full KS set.

1254: 

1255: \noindent Proposition p(2) from which in combination with p(1) the rules

1256: follow does not properly apply to individual ensemble-member values, realist

1257: possessed values $ \nu $,  and it is precisely this misapplication that is at the heart of the KS paradox.

1258: 

1259: \noindent The paradox may thus be understood from the following schematic:

1260: %

1261: \begin{center}

1262: $ \psi $  + p(2) $ \Rightarrow $ KS Paradox

1263: \end{center}

1264: %

1265: ($ \psi $ implying p(1) ). It is resolved by the replacement

1266: %

1267: \begin{center}

1268: $ \psi   \rightarrow  \Psi $

1269: \end{center}

1270: %

1271: Taken together then

1272: %

1273: \begin{equation}

1274: \left. \begin{array}{c} \Psi  + p(2) \\ {\rm or } \\ \psi  + p(1)

1275: \end{array}  \right\} \Rightarrow  \qquad {\rm no \; KS \; Paradox }

1276: \end{equation}

1277: %

1278: 

1279: \section{conclusion}

1280: 

1281: While quantum mechanics is generally accepted as the best tool available for

1282: the prediction of small-scale statistics, there remains disagreement over

1283: its physical interpretation particularly as regards its

1284: Hilbert-space state vectors. In the prevailing orthodox view $ \Psi $ is

1285: said to represent individual systems e.g. an individual electron of an

1286: ensemble.  It is understood in consequence that individual systems cannot

1287: consistently possess many of the common characteristics we

1288: attribute to them in ordinary language use (spatial position, mass, spin,

1289: etc.). The conceptual challenges presented by the view are obvious and

1290: well-known and an entire culture in science thinking has gone  to their

1291: explication \cite{112,104}. While this particular concept does indeed appear

1292: common to all orthodox-school thinking we caution not to overgeneralize

1293: with regard what is really a diversity of views \cite{140}.

1294: \noindent Likewise in characterizing the realist view \footnote{

1295: See refs.\cite[p. \ 44]{108}  and \cite[p. \ 9]{116}  } , particularly as

1296: impossibility proofs such as the ones here reviewed are directed against their

1297: very logic. Let us therefore carefully consider again the type of experimental

1298: arrangement cited earlier in illustration of realist thinking on the fundamental

1299: issue here, noncontextuality.

1300: 

1301: \noindent For the pair of illustrated spin-1 projection measurements

1302: %

1303: \begin{center}

1304: \scalebox{1.5}[1.5]{\includegraphics{fig10m.ps}}\\

1305: figure 17

1306: \end{center}

1307: %

1308: one always has the possessed-value relations

1309: %

1310: \bea

1311: \nu({\rm P }_1) + \nu({\rm P }_2) + \nu({\rm P }_3) &=& 1

1312: \eea

1313: %

1314: and

1315: %

1316: \bea

1317: \nu^\prime ({\rm P }_1^\prime) + \nu^\prime ({\rm P }_2^\prime) + \nu^\prime

1318: ({\rm P }_3^\prime) &=& 1

1319: \eea

1320: %

1321: We ask now whether in the realist view a result of measuring the projection $

1322: {\rm P }_2 = {\rm P }^\prime_2 $ may depend upon how the measurement is taken. Whether

1323: %

1324: \bea

1325: \nu ({\rm P }_2) &=& \nu^\prime ({\rm P }_2^\prime) \,  \label{1?}

1326: \eea

1327: %

1328: by necessity.

1329: Based on statements of the generality (e) above KS assumes that the realist

1330: view answers in the affirmative, the claim (not to beat a dead

1331: horse) at the heart of the derivation of their paradox, as shown in section

1332: 6.1.

1333: 

1334: \noindent But the question is not unique to experimental arrangements of the

1335: type pictured in figure (17). The very same also appears in Bell -Inequality

1336: (BI) analysis \cite{111} in the context of the question of space-time locality

1337: \footnote{ There, it is certainly true that what is to the realist view a question

1338: of contextuality (of the possible effects of $ {\rm D}_2 , \psi_2 , { \rm

1339: etc. \, \, on } \, \, \psi_1 $ , {\it assuming separability} - that the

1340: effects between paired particles are purely causal) is to the orthodox view

1341: one of separability only ( roughly, of the affects of decay particle $

1342: \psi_2 $ on particle $ \psi_1 $ intrinsic to their bound state ). Hence e.g.

1343: the realist issue with detector $ {\rm D}_2 $ triggering. }. There, a set of

1344: fixed projection measurements is taken (by detector $D_1$ at space-time

1345: position x ) on spin-1/2 particles, as the setting of a remote apparatus

1346: ($D_2$ at y ) is varied

1347: %

1348: \begin{center}

1349: \scalebox{1.5}[1.5]{\includegraphics{fig11m.ps}}\\

1350: figure 18

1351: \end{center}

1352: %

1353: One then considers contextuality question, whether

1354: %

1355: \bea

1356: \nu_\bk ({\rm P }_1) &=& \nu_\bk^\prime ({\rm P }_1)  \label{2?}

1357: \eea

1358: %

1359: where here too value subscripts and superscripts in obvious notation

1360: characterize the measurement context \footnote{ While it is possible for

1361: individual states and measurements $ (\psi, \; \nu ) $  to depend upon

1362: context while those of ensembles $ ( \Psi, \lan \; \ran ) $ remain

1363: independent, or noncontextual, Bell's inequality is concerned with ensemble

1364: contextuality (that of $ \Psi $ ) though by way of individual state

1365: contextuality (that of  $ \psi $ ). The KS paradox on the other hand treats the contextuality of individual states.}.

1366: 

1367: \noindent An affirmative answer without condition leads as is well-known

1368: to an realization of the Bell inequality and immediate disagreement of

1369: realism with observation. And no one has thought to do this. On the contrary,

1370: contextuality is assumed as a matter of course, its possible effect dismissed only

1371: upon satisfaction of the most stringent locality conditions

1372: %

1373: \bea

1374: ( x - y )^2 & > & 1 \label{3?}

1375: \eea

1376: %

1377: brought to some level of experimental realization in the work of

1378: Aspect et al and several refinements since \cite{110}. We are keen to

1379: make this point, as (to continue the beating) causal contextuality is indeed

1380: central to realist thinking particularly since the appearance of quantum

1381: phenomena and the realist conceptual involvement of hidden

1382: variables. D. Bohm confirms that " when we measure the momentum

1383: 'observable,' the final result is determined by hidden parameters in the

1384: momentum-measuring device as well as by hidden parameters in the observed

1385: electron" \cite{120}. N. Bohr also, himself no realist, agrees \footnote{

1386: "This crucial point... implies the impossibility of any sharp separation

1387: between the behavior of atomic objects and the interaction with the

1388: measuring instruments which serve to define the conditions under which the

1389: phenomena appear.", Ref. \cite[p. \ 209]{121} }, while others, more often

1390: than not on purely formalistic grounds, may not \cite{123}.  We conclude therefore that

1391: the value-definiteness of realism p(1) is in no way incompatible with causal

1392: contextuality not-p(2) but is on the contrary complemented by it \footnote{ Without naming it, as the term had yet

1393: to enter the language, Belinfante offers in part 1 of reference \cite{116}

1394: (up to page 78) an excellent insight into contextual thinking. Likewise Bell

1395: \cite{111}, the first to notice contextuality as an issue central

1396: to the KS paradox. There are many others.}.

1397: 

1398: \noindent If figure(18) by comparison with (17) (in light of constraint

1399: (\ref{3?})) is taken to indicate an experimental constraint appropriate to noncontextuality in the KS analysis, the feasibility of such a test seems

1400: clearly in doubt. We will not persue the question here.

1401: 

1402: \noindent What is striking about the realist view of possessed values is its

1403: marked lack of novelty. The possessed values of an electron are thought to

1404: be no different in their manner of possession than tose of tables, trees,

1405: cats; a stone said to possess 10 kilograms is said in the same breath to

1406: weigh 98 Newtons, to oscillate with a frequency $ f =  \frac{1}{2\pi }

1407: \surd\overline{\frac{k}{10}} $ \, when attached to a spring k, and so

1408: forth; i.e., such a stone possesses all the values that a stone of 10

1409: kilograms might yield upon measurement, many of which from the laws of mechanics may be known before hand.... To say that the stone is 10 kilograms is a manner of speaking, a shorthand for an otherwise

1410: unwieldy usage.

1411: 

1412: \noindent In absence such a shorthand however one is left with the

1413: directly observed or predicted values themselves, no

1414: less possessed by the object. In this form they may well appear in some ways

1415: peculiar and to have unusual properties \footnote{ van Fraassen's

1416: ontological contextualism has taken special ridicule for its proliferation

1417: of possessed values that e.g. "pop in and out of existence" at the whim of

1418: the experimentalist \cite{107}, "de-Ockhamizing QM!" \cite[p. \ 135]{108} .

1419: Such a criticism on mathematical form, Bell might have countered, merely

1420: betrays a lack of sufficient imagination \cite[p. \ 64]{200}. Einstein too

1421: was largely unconcerned with formal structures per se \cite[pp. \

1422: 234-238]{104} . } . They do not; and the appearance generally dissolves upon

1423: careful consideration: Again, the manner of microscopic property possession in this view is no different than that for macroscopic possession.

1424: 

1425: \noindent It has long been considered not possible to assign definite values to

1426: systems whose possessed values are described by noncommuting QM operators.

1427: The KS paradox proves this for a discrete set of observables and as such

1428: is a statement of the limitation of QM to predict individual microscopic phenomena. But the paradox is not a proper criticism

1429: of realist thinking; it reduces to absurdity a view held by no one.

1430: 

1431: \vspace{10mm}

1432: 

1433: \begin{thebibliography}{}

1434: 

1435: 

1436: \bibitem{101}

1437: C. Cohen-Tannoudji, B. Diu, and F. Laloe, {\sl  Quantum Mechanics }, New

1438: York: John Wiley \& Sons (1977).

1439: 

1440: \bibitem{103}

1441: B. d'Espagnat, {\sl Conceptual foundations of Quantum Mechanics}, Menlo

1442: Park, California: W. A. Benjamin, Inc. (1971).

1443: 

1444: \bibitem{102}

1445: F. Mosteller, R. Rourke, and G. Thomas, { \it Probability and Statistics},

1446: London: Addison-Wesley (1961).

1447: 

1448: \bibitem{125}

1449: D. Brody and L. Hughston, J. Math. Phys. {\bf 43 }, 5254 (2002); Stanford

1450: Encyclopedia of Philosophy, {\sl Collapse Theories },

1451: http://plato.stanford.edu/entries/qm-collapse/ .

1452: 

1453: \bibitem{122}

1454: B.C. van Fraassen, 'The Problem of Indistinguishable Particles', in E.

1455: Castellani (ed.), {\sl Interpreting Bodies }, 73, Princeton: Princeton

1456: University Press (1998); Stanford Encyclopedia of Philosophy, {\sl The

1457: Identity of Indiscernibles  },

1458: http://plato.stanford.edu/entries/identity-indiscernible/ (May, 2002).

1459: 

1460: 

1461: \bibitem{116}

1462: F. J. Belinfante, {\sl A Survey of Hidden-Variables Theories }, Oxford, New

1463: York: Pergamon Press (1973).

1464: 

1465: \bibitem{104}

1466: M. Jammer, {\sl The Philosophy of Quantum Mechanics }, New York: John Wiley

1467: \& Sons, Inc. (1974).

1468: 

1469: 

1470: \bibitem{118}

1471: P. Mittelstaedt, {\sl The Interpretation of Quantum Mechanics and the

1472: Measurement Process }, Cambridge: Cambridge University Press (1998).

1473: 

1474: \bibitem{105}

1475: A. Einstein, B. Podolsky, and N. Rosen, Phys. Rev., {\bf 47 }, 777 (May

1476: 1935).

1477: 

1478: \bibitem{5}

1479: J.S.Bell, Physics, {\bf 1}, 195 (1964).

1480: 

1481: \bibitem{111}

1482: J. S. Bell, Rev. Mod. Phys., {\bf 38 }, 447 (1966). Reprinted in {\sl

1483: Speakable and unspeakable in quantum mechanics }, Cambridge: Cambridge

1484: University Press (1987).

1485: 

1486: \bibitem{124}

1487: F. Selleri, 'Realism and the Wave-Function of Quantum Mechanics', in B.

1488: d'Espagnat (ed.), International School of Physics, {\sl Foundations of

1489: Quantum Mechanics }, 398, New York: Academic Press (1971).

1490: 

1491: \bibitem{106}

1492: S. Kochen and E. Specker, J. Math. \& Mech., {\bf 17 }, 59 (1967 ).

1493: 

1494: \bibitem{107}

1495: Stanford Encyclopedia of Philosophy, The Kochen-Specker Theorem,

1496: http://plato.stanford.edu/entries/kochen-specker/index.html (Sep., 2000).

1497: 

1498: \bibitem{108}

1499: M.L.G. Redhead, {\sl Incompleteness, Nonlocality, and Realism }, Oxford:

1500: Clarendon Press (1987).

1501: 

1502: \bibitem{109}

1503: J. von Neumann, {\sl Mathematical Foundations of Quantum Mechanics },

1504: Princeton: Princeton University Press (1955), p. 295-328;

1505: J. Albertson, Am. J. Phys., {\bf 29 }, 478 (1961); Ref. \cite[pp. \

1506: 265-269]{104}; L. E. Ballentine, Rev. Mod. Phys. {\bf 42}, 358 (1970).

1507: 

1508: \bibitem{114}

1509: There seems to be, according to \cite{115}, some indication of this in J.

1510: Tharrats, Comptes Rendus {\bf  250}, 3786 (1960), with which, by the way

1511: Clifton in \cite{115} emphatically disagrees. But I have not found the

1512: English translation.

1513: 

1514: \bibitem{113}

1515: A. M. Gleason, J. Math. \& Mech., {\bf 6 }, 885 (1957 ); Ref. \cite[pp. \

1516: 265-269]{104}; Ref. \cite[pp. \ 27-30]{108}; Ref. \cite{111}.

1517: 

1518: \bibitem{119}

1519: F. D. Peat, {\sl Einstein's Moon }, Chicago: Contemporary Books (1990).

1520: 

1521: \bibitem{117}

1522: B. Van Fraassen, 'Semantic Analysis of Quantum Logic', in C. A. Hooker

1523: (ed.), {\sl Contemporary Research in the Foundations and Philosophy of

1524: Quantum Theory }, 80, Dordrecht: Reidel (1973).

1525: 

1526: \bibitem{112}

1527: K.Pooper, {\sl Quantum Theory and the Schism in Physics }, London:

1528: Hutchinson \& Co. (1982).

1529: 

1530: \bibitem{110}

1531: A.Aspect, Phys.\ Rev. {\bf D14}, 1944 (1976); E.Wigner, Am. J. Phys., {\bf

1532: 33 }, 1005 (1970).

1533: 

1534: \bibitem{140}

1535: See introduction to R. Healey's, {\sl The philosophy of quantum mechanics },

1536: Cambridge: Cambridge University Press (1989).

1537: 

1538: \bibitem{120}

1539: D. Bohm, Phys. Rev., {\bf 85}, 180 (1952).

1540: 

1541: \bibitem{121}

1542: N. Bohr, 'Discussion with Einstein on Epistemological Problems in Atomic

1543: Physics', in P.A. Schilp (ed.), Library of Living Philosophers, {\sl A.

1544: Einstein: Philosopher Scientist }, 201, New York: Tudor (1951).

1545: 

1546: \bibitem{123}

1547: See e.g. 'Concluding Remarks' in, B. Misra, Nuovo Cimento {\bf 47 }, 841

1548: (1966).

1549: 

1550: \bibitem{115}

1551: R. Clifton, Am. J. Phys., {\bf 61 }, 443 (1993).

1552: 

1553: \bibitem{200}

1554: J. Mehra, {\sl Quantum Principle: its interpretation and epistemology},

1555: Dordrecht: Reidel (1974).

1556: 

1557: \end{thebibliography}

1558: 

1559: \end{document}

1560: 

1561: 

1562: 

1563: 

1564: 

1565: 

1566: 

1567: